Guest contributor David Robertson discusses the differences between risk simulations and real risk data in relation to quad bike safety:
“The motion picture blockbuster “Avatar”, for the time we are in the cinema, would have us believe that Sam Worthington can turn into a giant blue man on a faraway planet. James Cameron used computers to deliver us this illusion. In science (and particularly with safety) we must be able to distinguish between computers that are valuable tools and computers that don’t represent reality. Dynamic Research, Inc.(DRI) chose 113 actual quad bike (ATV) accidents to simulate in a computer model.
The first table below shows the injuries DRI record as what really happened from all the 113 cases, but represented as approximate normalized injury cost (ANIC)*. When DRI’s computer runs the same cases, one would expect a similar result. The results are shown in the second table (ANIC), note the scale on the left has had to be changed (table 2 should be about 10 times taller if the same scale in table 1 was used) because head injuries rose from 360 points to well over 3000. Equally astonishingly is that abdomen injuries have vanished altogether and chest injuries dropped from 266 to a next to nothing (23). Asphyxiation was not even included in the computer model.
Remembering that computer modelling is only a tool. It is a fundamental part of science to establish your tools are accurate and representative, if they are not they must be tuned until they are or disposed. The two tables are light-years apart in their results depicting the real life cases and DRI’s computer simulation and therefore DRI’s computer simulation can’t be relied on.
The question for some that are unfamiliar with computer simulations is why? The short answer is, were DRI to plug a solution into their computer, say a “helmet”, the computer is very “happy” because you have dealt with the over-exaggerated problem of head injury. But if DRI’s computer is asked to deal with a rollover protection device it gets very “angry” and says don’t be silly, there is no problem with chest, abdomen or asphyxiation. I it doesn’t know any better.
The fact is that it is more than probable that any safety system that is not very, very effective in dealing with head injuries is doomed to failure if assessed by DRI”s computer.”
*For clarity only body regions and causes where fatalities were recorded are displayed (leg, face and neck injuries are excluded).